Pylon vs. Firebase

Firebase is Google's mobile-first BaaS — Firestore, Realtime Database, Auth, Cloud Functions, Storage. Pylon overlaps that surface and adds declarative schema, faceted search, and authoritative game shards.

Start free on Pylon Cloud →Read the docs →

TL;DR

Pick the one that fits your shape

Choose Firebaseif

You're building a mobile-first app, you want Google's ecosystem (FCM push, Crashlytics, GA4) tightly integrated, and you accept a closed-source backend.

Choose Pylonif

You want self-host, MIT/Apache license, declarative schema instead of schemaless documents, faceted search without Algolia, no cold starts on functions, or game shards.

Architecture

Where the two diverge

	Pylon	Firebase
License	MIT OR Apache-2.0	Closed source
Self-hostable	Yes — any Linux box	No — Google-only
Schema	Declarative (TypeScript)	Schemaless (Firestore documents)
Functions runtime	Bun, in-process	Cloud Functions (Node, separate)
Function cold start	None	1–10 seconds for cold containers
Full-text search	FTS5 built-in	Mirror to Algolia / Typesense / Meilisearch
Game shards	Yes (`Shard<S: SimState>`)	No

Same shape

What both ship

Either one is a real choice for a managed mobile-first backend. The differences below are about emphasis and operational shape, not feature presence.

Real-time sync over WebSocket
Built-in auth (email, OAuth, anonymous)
Functions for server-side logic
File storage with signed URLs
Web + mobile + native SDKs
Managed cloud (Firebase / Pylon Cloud)

Where Pylon wins

What you get with Pylon you don't with Firebase

Declarative schema, not schemaless documents

Firestore lets any document have any shape. Great for prototyping, painful at scale (typo'd field names become orphan rows; refactors are manual sweeps). Pylon's entity() definition is the single source of truth.

No cold starts

Pylon functions run in-process. Latency is microseconds. Firebase Cloud Functions' cold-start tax (1–10s) is the single most-complained-about thing about Firebase.

Functions share a transaction with writes

ctx.db inside a mutation is a real transaction. Firebase Cloud Functions hit Firestore via the Admin SDK over network — separate process, no atomicity.

Full-text + faceted search built-in

FTS5 + roaring-bitmap facets in the binary. Firebase explicitly tells you to integrate Algolia for search. No third-party bill, no second system to keep in sync.

Self-host on day one

A Linux box and pylon serve. No vendor outage takes your app down. Firebase is Google-only — when Google's status page is red, you're red.

Predictable pricing

Firebase pricing is famously hard to reason about — reads, writes, deletes, egress, function invocations, GB-seconds, all priced separately. Pylon Cloud is one number per dimension. Self-hosted, you pay your VPS bill.

Where Firebase wins

What Firebase does better today

Honest comparison — these are real reasons to pick Firebase. If any of them are dealbreakers, choose accordingly.

Google ecosystem integrations

FCM push notifications, Crashlytics, GA4, Remote Config, A/B testing, in-app messaging. Mobile-app-shaped concerns are first-class. Pylon focuses on the backend; for these, you'd integrate FCM + Sentry + your analytics provider on top.

Mobile-first SDKs

Firebase's iOS, Android, and Unity SDKs are deeply integrated with platform features (push tokens, app-startup hooks, offline persistence). Pylon's mobile SDKs are first-class but younger.

Unlimited horizontal scale by design

Firestore is built to shard transparently. Pylon's SQLite default tops out around 70k writes/sec single-process; Postgres mode scales further but eventually hits limits. For Twitter-scale apps both Pylon and SQLite are wrong — you want Spanner / DynamoDB / Cassandra.

Migration

Coming from Firebase

Most of the dev surface translates one-to-one. The biggest deltas show up as differences in shape, not features missing.

Firebase	Pylon
Firestore collections	`entity()` definitions in TypeScript
Security rules (custom DSL)	`policy()` with boolean expressions
`firestore.collection().onSnapshot()`	`useQuery("Entity")`
Cloud Functions (Node, HTTP-triggered)	`mutation` / `action` in `functions/*.ts`
Firebase Auth	Magic codes / password / OAuth (export users, import as `User` rows)
Cloud Storage	`/api/files/init` → direct PUT (Stack0 / S3 / local)
Firebase Cloud Messaging	Keep FCM — register tokens via a Pylon `action`, send from your function

Honest weakness

Where Pylon loses

Firebase's mobile-first integrations (FCM, Crashlytics, A/B testing via Remote Config, in-app messaging) have no Pylon equivalent. If your app's most important behaviors are around push notifications and mobile experimentation, Firebase has more polish. Pylon assumes you'll bring your own analytics + crash reporting + push provider.

Both / and

When using both is the right call

Use Pylon for backend logic (data, sync, functions, auth) and FCM for push notifications. They don't conflict — push is a lightweight integration, not a stack commitment. Many teams ship this combination to get Firebase-quality push without locking the rest of the stack into Google.

Try Pylon — free Hobby tier on Cloud

No card, no setup. Run a real Pylon project against managed Postgres in under a minute. Migrate from Firebasewhen you're ready — or run both.

Start free on Pylon Cloud →